sebi wrote:It also has nothing to do with tcgplayer or with trying to monetize the website, i'm not sure how you got to that conclusion.

I'm not sure how you got to yours, either. I wasn't referring to monetizing the website.

sebi wrote:Like in any kind of partly protected system, we here want to protect our users from abuse. In that case 20 people were just about to lose 10$ each, due to a fluctuation that deckbox prices did not respond to (so a deckbox bug you can say), and a guy who thought he's entitled to steal that value because he noticed the bug. Also he did not bother to tell us about it when he saw it, instead he opened 20 trades with people who did not know the price spiked 2 days before.

The case referenced before is not about marginal arbitrage. The card was mistakenly marked on deckbox as being three times less expensive than 90% of all other websites - ebay, tcgplayer, amazon, star city games, adventureson, all of them, with a 10$ difference on a 5$ card.

I'm not yet sure what the solution to this problem is, or how to phrase a rule that clearly explains what is abuse. But I still stand very decided that arbitrage of that sort on market fluctuations is pretty clearly abuse. Preventing that is of course not an easy problem to solve, but I would still like to prevent it. Until we have a good solution, suspending people who blatantly do it is a temporary one.

Feedback is as always welcome on the topic. I do not want to ignore the problem.

Also interesting points. Let me throw this scenario at you so you can tell me what Deckbox's stance would be in that case:
When I'm interested in a card, I'll often initiate trades with multiple users. The reasons are several, including, but not limited to, a) I want several copies of the card, b) I know many users on Deckbox either ignore trades (maybe they have too many to look through, I don't know) or cancel them (often without a word), c) several trade negotiations will likely break down.

With your above explanation, my understanding is if that card happens to be one experiencing such a glitch and I don't realize it, I'll be tried/judged as a scammer. The conclusions I take from this are:

• the person suggesting a card (I assume both when initiating a trade or during trade negotiations) is responsible for checking other market prices (presumably TCGmid) on the other person's behalf to ensure the price isn't experiencing a price fluctuation
• or, limit the number of trades any given card is in to avoid being accused as a scammer
• the burden of proof/reputation liability lies with whomever suggests a card for trade

Thoughts?

renoan wrote:

sebi wrote:It also has nothing to do with tcgplayer or with trying to monetize the website, i'm not sure how you got to that conclusion.

I'm not sure how you got to yours, either. I wasn't referring to monetizing the website.

sebi wrote:Like in any kind of partly protected system, we here want to protect our users from abuse. In that case 20 people were just about to lose 10$ each, due to a fluctuation that deckbox prices did not respond to (so a deckbox bug you can say), and a guy who thought he's entitled to steal that value because he noticed the bug. Also he did not bother to tell us about it when he saw it, instead he opened 20 trades with people who did not know the price spiked 2 days before.

The case referenced before is not about marginal arbitrage. The card was mistakenly marked on deckbox as being three times less expensive than 90% of all other websites - ebay, tcgplayer, amazon, star city games, adventureson, all of them, with a 10$ difference on a 5$ card.

I'm not yet sure what the solution to this problem is, or how to phrase a rule that clearly explains what is abuse. But I still stand very decided that arbitrage of that sort on market fluctuations is pretty clearly abuse. Preventing that is of course not an easy problem to solve, but I would still like to prevent it. Until we have a good solution, suspending people who blatantly do it is a temporary one.

Feedback is as always welcome on the topic. I do not want to ignore the problem.

Also interesting points. Let me throw this scenario at you so you can tell me what Deckbox's stance would be in that case:
When I'm interested in a card, I'll often initiate trades with multiple users. The reasons are several, including, but not limited to, a) I want several copies of the card, b) I know many users on Deckbox either ignore trades (maybe they have too many to look through, I don't know) or cancel them (often without a word), c) several trade negotiations will likely break down.

With your above explanation, my understanding is if that card happens to be one experiencing such a glitch and I don't realize it, I'll be tried/judged as a scammer. The conclusions I take from this are:

• the person suggesting a card (I assume both when initiating a trade or during trade negotiations) is responsible for checking other market prices (presumably TCGmid) on the other person's behalf to ensure the price isn't experiencing a price fluctuation
• or, limit the number of trades any given card is in to avoid being accused as a scammer
• the burden of proof/reputation liability lies with whomever suggests a card for trade

Thoughts?

Come on. Yes, for any suspicious usage pattern we could generate a hypothetical where that usage pattern would actually be the result of innocent behavior, like the case where four people ship packages to me and they all coincidentally get lost and it isn't my fault, but patterns are taken as evidence in these cases and even if they can generate false positives. So yes, you can be tried/judged incorrectly under extremely-unlikely circumstances, but what does this prove? The guy involved with the Polukranos case didn't even try this argument since it was so implausible (notably, he was *only* targeting the version of the card that was underpriced in trades - that's what made it obvious to begin with!)

What you're doing isn't showing that there's a problem with the rule, but there's a problem with any fact-finding process that takes patterns as evidence. You're proving way too much here.

9700377 wrote:

elpablo wrote:I think the deck box team needs to do their best to get prices right, but there needs to be a disclaimer somewhere that says basically 'hey, we're not responsible if you let yourself get boned".

Or they can do what they're doing - do their best, but when a bug exists and is exploited, ban the exploiter. It really doesn't upset the balance that much, and it helps built trust in Deckbox's pricing system, which is important. Like, I have a ton of cards listed for sale and I peg all the prices to the Deckbox price because it's too much work to maintain them on my own. If there's a pricing error that someone takes advantage of, I can cancel those sales and report the user and not have to worry. The alternative - that I have to manually monitor all my prices or just not list cards because the downside of getting fucked even once can wipe out the return from 10+ sales - is clearly unacceptable.

I don't think the lines are that hard to draw here. Yes, there are users who purposely target cards that are listed at lower values on Deckbox than other markets and that's fine if those prices aren't the result of clear bugs or anomalies in the pricing system. The Polukranos case got attention because it was obvious. afaik there haven't been any other users who have gotten in hot water over this.

Gotta tell you, this makes me super nervous about buying from anyone on Deckbox lest they take a similar stance to you. I don't always price shop for cards since I usually don't really know how much a card is worth. Glittering Wish, for example, to me is worth $2, maybe, regardless of what market hype has done to the price. So had I not know the price spiked over the last month or so, and I'd seen a $2 price tag I would nod my head in agreement and think, "Yeah, I suppose I could do that. Give me a playset."

That would get reported by you and cause me at the least grief if not a banning. After all, how do I prove in the court of public opinion that I didn't know Glittering Wish went up in price?

9700377 wrote:Come on. Yes, for any suspicious usage pattern we could generate a hypothetical where that usage pattern would actually be the result of innocent behavior, like the case where four people ship packages to me and they all coincidentally get lost and it isn't my fault, but patterns are taken as evidence in these cases and even if they can generate false positives.

Patterns are circumstantial evidence. The line of thinking you outline is dangerously close to "guilty until proven innocent."

*EDIT*
On an unrelated note, I just noticed that my registration date shows 2013 and not 2014. How'd that happen?

renoan wrote:

9700377 wrote:

elpablo wrote:I think the deck box team needs to do their best to get prices right, but there needs to be a disclaimer somewhere that says basically 'hey, we're not responsible if you let yourself get boned".

Or they can do what they're doing - do their best, but when a bug exists and is exploited, ban the exploiter. It really doesn't upset the balance that much, and it helps built trust in Deckbox's pricing system, which is important. Like, I have a ton of cards listed for sale and I peg all the prices to the Deckbox price because it's too much work to maintain them on my own. If there's a pricing error that someone takes advantage of, I can cancel those sales and report the user and not have to worry. The alternative - that I have to manually monitor all my prices or just not list cards because the downside of getting fucked even once can wipe out the return from 10+ sales - is clearly unacceptable.

I don't think the lines are that hard to draw here. Yes, there are users who purposely target cards that are listed at lower values on Deckbox than other markets and that's fine if those prices aren't the result of clear bugs or anomalies in the pricing system. The Polukranos case got attention because it was obvious. afaik there haven't been any other users who have gotten in hot water over this.

Gotta tell you, this makes me super nervous about buying from anyone on Deckbox lest they take a similar stance to you. I don't always price shop for cards since I usually don't really know how much a card is worth. Glittering Wish, for example, to me is worth $2, maybe, regardless of what market hype has done to the price. So had I not know the price spiked over the last month or so, and I'd seen a $2 price tag I would nod my head in agreement and think, "Yeah, I suppose I could do that. Give me a playset."

That would get reported by you and cause me at the least grief if not a banning. After all, how do I prove in the court of public opinion that I didn't know Glittering Wish went up in price?

If the price tag was $2 because I had failed to update for the past month, my bad. If it was $2 because the Deckbox pricing algorithm was off, then yes I would cancel. Selling my wishes for $2 would cause me much more grief than you getting your order canceled. But it seems unreasonable to claim this would make you nervous about buying from *anyone* - you should only be nervous about buying under unusual circumstances where the Deckbox price is way off. This is in fact uncommon!

And that's part of the reply to the points others have made about how these protections aren't afforded on other sites - firstly, I don't know how TCGPlayer works but I imagine that its algorithm is considered more-infallible. Can merchants peg their prices to the TCG prices? If one of these prices gets messed up, does TCGPlayer tells the merchants "too bad"? Show me a case of this - I'm genuinely curious. But even so, one *major* difference here is that TCGPlayer's merchants are much more-likely to be brick and mortar establishments that don't need to reply on a price index like people like me do in order to market a large number of cards. That is, the userbase of Deckbox is qualitatively-different from that of other sites in ways that justifies additional protection for merchants that may not exist elsewhere. Look at the top sellers list of the site - out of the top 25 or so, only one is a certified store! Deckbox should not only be accepting these differences, but leveraging them to profit from a market niche that can be inhabited by sellers like me.

Patterns are circumstantial evidence. The line of thinking you outline is dangerously close to "guilty until proven innocent."

Sufficiently-strong circumstantial evidence can satisfy any statistically-defined evidentiary burden. Again, let's zoom back to how this has implicit rule has been actually applied: One case where the user didn't even feign innocence. You simply can't look at the record and say that the risk of false convictions seems too high. Busting out wild thought experiments and hypotheticals is an unproductive exercise given the obvious conservatism with which this rule is being applied.

9700377 wrote:If the price tag was $2 because I had failed to update for the past month, my bad. If it was $2 because the Deckbox pricing algorithm was off, then yes I would cancel. Selling my wishes for $2 would cause me much more grief than you getting your order canceled.

Can merchants peg their prices to the TCG prices? If one of these prices gets messed up, does TCGPlayer tells the merchants "too bad"? Show me a case of this - I'm genuinely curious.

You misunderstand. Apologies for not being clearer. I'm 100% with you that you have the right to cancel the sale (I would, too) and report the price problem to Deckbox. My issue was with your statement that you would also report the user attempting to buy from you.

9700377 wrote:But even so, one *major* difference here is that TCGPlayer's merchants are much more-likely to be brick and mortar establishments that don't need to reply on a price index like people like me do in order to market a large number of cards. That is, the userbase of Deckbox is qualitatively-different from that of other sites in ways that justifies additional protection for merchants that may not exist elsewhere. Look at the top sellers list of the site - out of the top 25 or so, only one is a certified store!

A great point. There are inherent risks involved when going into business (essentially what a person does when putting cards up for sale) and any help Deckbox can give via tools or bettered algorithms is a welcome move.

9700377 wrote:

Patterns are circumstantial evidence. The line of thinking you outline is dangerously close to "guilty until proven innocent."

Sufficiently-strong circumstantial evidence can satisfy any statistically-defined evidentiary burden.

It doesn't satisfy "without a shadow of a doubt." This is a separate topic, but it's a worrisome development in the American justice system.

renoan wrote:

9700377 wrote:If the price tag was $2 because I had failed to update for the past month, my bad. If it was $2 because the Deckbox pricing algorithm was off, then yes I would cancel. Selling my wishes for $2 would cause me much more grief than you getting your order canceled.

Can merchants peg their prices to the TCG prices? If one of these prices gets messed up, does TCGPlayer tells the merchants "too bad"? Show me a case of this - I'm genuinely curious.

You misunderstand. Apologies for not being clearer. I'm 100% with you that you have the right to cancel the sale (I would, too) and report the price problem to Deckbox. My issue was with your statement that you would also report the user attempting to buy from you.

I think if it turned out you had gone around and bought every copy of Glittering Wish available on the site, and it could be reasonably demonstrated that the low prices you bought at were the result of an error in the pricing then reporting you would be justified. To me the important question would be whether you knowingly exploited the pricing, which may be difficult to show in some cases - for example, if the price was low but then corrected but people still had it automatically pegged to the low price and hadn't updated, then one could buy out those cards without realizing that the prices were the result of a Deckbox hiccup and in that case I don't think you should be banned. And yes, if these cases seemed to occur often and people were getting investigated spuriously I think this would be a problem for Deckbox. But again, this was not the case in the Polukranos example and no cases of this occurring have arisen, therefore I don't think worrying about this can be justified.

If the price tag was $2 because I had failed to update for the past month, my bad. If it was $2 because the Deckbox pricing algorithm was off, then yes I would cancel. Selling my wishes for $2 would cause me much more grief than you getting your order canceled. But it seems unreasonable to claim this would make you nervous about buying from *anyone* - you should only be nervous about buying under unusual circumstances where the Deckbox price is way off. This is in fact uncommon!

canceling orders because the price goes up/down/is wrong is the same shady crap people do on TCG player that gets them flamed on reddit :P

it's [u]your [/u]responsibility to make sure [u]your [/u]prices are correct.

elpablo wrote:

If the price tag was $2 because I had failed to update for the past month, my bad. If it was $2 because the Deckbox pricing algorithm was off, then yes I would cancel. Selling my wishes for $2 would cause me much more grief than you getting your order canceled. But it seems unreasonable to claim this would make you nervous about buying from *anyone* - you should only be nervous about buying under unusual circumstances where the Deckbox price is way off. This is in fact uncommon!

canceling orders because the price goes up/down/is wrong is the same shady crap people do on TCG player that gets them flamed on reddit :P

it's [u]your [/u]responsibility to make sure [u]your [/u]prices are correct.

As a positive statement, no it's not my responsibility. If Deckbox's algorithm egregiously misprices a card (and I don't mean merely that it's a bit higher or lower than I'd like, or it's a few hours behind a spike), and I peg my prices to that algorithm and it gets bought out, then my understanding is that I can cancel the order. Period. This is *not* similar to the cases that get called out on Reddit, which are primarily concerned with shit like "Rabblemaster spiked from $1 to $5 over the weekend and now my order was canceled" stuff.

9700377 wrote:Again, let's zoom back to how this has implicit rule has been actually applied: One case where the user didn't even feign innocence. You simply can't look at the record and say that the risk of false convictions seems too high. Busting out wild thought experiments and hypotheticals is an unproductive exercise given the obvious conservatism with which this rule is being applied.

Oops, didn't read this part.

I see your point in wanting to take an overall look at the issue and wanting to make a "what's best for the community" decision. I simply disagree with you in your assessment that the false conviction rate is low enough to be acceptable. I have no idea how high or low the false conviction rate is at the moment, but what I do know is that any system should strive for a 0 false conviction rate. We may only be talking about card trading here and not an actual justice system, but the principles are the same (regarding this particular discussion).

To reel this back a little, the point of this whole discussion is to determine how to deal with situations where a person is involved in a number of trades that all involve a given card that is seeing an out-of-the-ordinary price difference as compared to other markets, correct? I'm arguing it's each person's responsibility to determine his or her own perception of value for a card, regardless of what the pricing shows, and those arguing otherwise are saying it's not; the responsibility lies with the person who proposes a given card in a trade.

Regardless of the initiator's intentions, those arguing for a protectionist environment for card owners I think agree with me that pricing fluctuations are an instigator to the problem. I don't know that background, but jassi earlier said:

jassi007 wrote:However in the process of doing that, they lost their price data from tcgplayer. Basically TCGP saw them as competition, and no longer allowed them to import price data.

Assuming that's right, I'm guessing importing price data from any site that is dedicated to selling wouldn't fly. So can a neutral site be used instead? MTGGoldfish and MTGPrice come to mind.

renoan wrote: So can a neutral site be used instead? MTGGoldfish and MTGPrice come to mind.

MTGGoldfish uses TCGPlayer pricing data so that isn't an option.

MTGPrice uses aggregated data from several different sites including TCG, Amazon, CFB, etc. That might be an option.

renoan wrote:I see your point in wanting to take an overall look at the issue and wanting to make a "what's best for the community" decision. I simply disagree with you in your assessment that the false conviction rate is low enough to be acceptable. I have no idea how high or low the false conviction rate is at the moment, but what I do know is that any system should strive for a 0 false conviction rate.

It's 0%. 1 person was convincted, and he did not even say he is innocent. Evidence was obvious as well. That's all the convictions in the last 6 years of deckbox :)

sebi wrote:

renoan wrote:I see your point in wanting to take an overall look at the issue and wanting to make a "what's best for the community" decision. I simply disagree with you in your assessment that the false conviction rate is low enough to be acceptable. I have no idea how high or low the false conviction rate is at the moment, but what I do know is that any system should strive for a 0 false conviction rate.

It's 0%. 1 person was convincted, and he did not even say he is innocent. Evidence was obvious as well. That's all the convictions in the last 6 years of deckbox :)

In fact, he said he was guilty, it wasn't just an absence of declaring innocence. It really rubs me the wrong way when someone says they are fully aware they are doing something unethical (which is what that guy said), but it's not against some set of rules so that makes it ok. Have a little more sense of self worth, if you think it's unethical then don't do it, period.

sebi wrote:

renoan wrote:I see your point in wanting to take an overall look at the issue and wanting to make a "what's best for the community" decision. I simply disagree with you in your assessment that the false conviction rate is low enough to be acceptable. I have no idea how high or low the false conviction rate is at the moment, but what I do know is that any system should strive for a 0 false conviction rate.

It's 0%. 1 person was convincted, and he did not even say he is innocent. Evidence was obvious as well. That's all the convictions in the last 6 years of deckbox :)

Awesome!

bactgudz wrote:It really rubs me the wrong way when someone says they are fully aware they are doing something unethical (which is what that guy said), but it's not against some set of rules so that makes it ok. Have a little more sense of self worth, if you think it's unethical then don't do it, period.

Problem is, ethics are a gray area. Whether something is "unethical" or not is a judgment call, but whether it goes against a written rule is rather clear. As I said in the other thread, you can't punish someone just for having different morals than you.

Personally, I see only three really good reasons to cancel a sale on this or any other website.

One: went to pull the card, and had no copies/not enough copies to fill that order. In that case, there should be a specific button to click that takes the seller to a page that you type the quantity you are short by. Once you click OK on that page, it takes the seller to another page with a script that automatically unlists that card from their sale list, if they have no copies at all, or reduces the sell list by that number of copies, and locks them out of adding that card back to their sell list under any price for 24 hours, minimum, to prevent users from abusing the cancel order button. It would also generate an email to the buyer/s to either let them know their order was cancelled, or they'd be receiving fewer copies of the card, and they have to confirm whether or not they still want the order.

Two: there should be a button that says: "report suspected deckbox pricing error/issue with order".This would generate a hold on the order, email Sebi or another admin of the issue, and give them the chance to look at the report, judge whether there is an error, and confirm or deny the cancellation. If they deny it, the order comes off hold, and you are on the hook for shipping the card anyway. It would also give you the opportunity to alternately send an admin a bug report/other cancellation reason, such as customer on ban list or suspected scammer, or even "outside of seller's shipping limits. The number of times you use this cancellation method would be recorded, as would the number of times the admins judged your use as valid, and would be used to whether you're abusing the button or not.

Three, a button that says "family emergency". This button would cancel all pending orders that had not been paid, would put your account on an automatic minimum two to three day, although you could specify longer, trading, buying, and selling suspension, similar to vacation mode, but more broad in scope.

These three suggestions have extreme effects, obviously, but they would also ensure that people are only using them when absolutely necessary, as abuse would disrupt the abuser's use of the site. If you don't have the card, then, barring buying the card, why would you be relisting it for sale right away? Are you clicking the pricing error issue because there's an actual issue, or because the price of the card shot up overnight, and you want to back out? Is there a real family emergency? If so, you won't mind suspending all sales for a day or two anyway, to take care of your family or recover from your issue.